Robust automatic face clustering in news video

Clustering identities in a video is a useful task to aid in video search, annotation and retrieval, and cast identification. However, reliably clustering faces across multiple videos is challenging task due to variations in the appearance of the faces, as videos are captured in an uncontrolled environment. A person's appearance may vary due to session variations including: lighting and background changes, occlusions, changes in expression and make up. In this paper we propose the novel Local Total Variability Modelling (Local TVM) approach to cluster faces across a news video corpus; and incorporate this into a novel two stage video clustering system. We first cluster faces within a single video using colour, spatial and temporal cues; after which we use face track modelling and hierarchical agglomerative clustering to cluster faces across the entire corpus. We compare different face recognition approaches within this framework. Experiments on a news video database show that the Local TVM technique is able effectively model the session variation observed in the data, resulting in improved clustering performance, with much greater computational efficiency than other methods.

Semantic Classes in Topic Detection and Tracking

Topic detection and tracking (TDT) is an area of information retrieval research the focus of which revolves around news events. The problems TDT deals with relate to segmenting news text into cohesive stories, detecting something new, previously unreported, tracking the development of a previously reported event, and grouping together news that discuss the same event. The performance of the traditional information retrieval techniques based on full-text similarity has remained inadequate for online production systems. It has been difficult to make the distinction between same and similar events. In this work, we explore ways of representing and comparing news documents in order to detect new events and track their development. First, however, we put forward a conceptual analysis of the notions of topic and event. The purpose is to clarify the terminology and align it with the process of news-making and the tradition of story-telling. Second, we present a framework for document similarity that is based on semantic classes, i.e., groups of words with similar meaning. We adopt people, organizations, and locations as semantic classes in addition to general terms. As each semantic class can be assigned its own similarity measure, document similarity can make use of ontologies, e.g., geographical taxonomies. The documents are compared class-wise, and the outcome is a weighted combination of class-wise similarities. Third, we incorporate temporal information into document similarity. We formalize the natural language temporal expressions occurring in the text, and use them to anchor the rest of the terms onto the time-line. Upon comparing documents for event-based similarity, we look not only at matching terms, but also how near their anchors are on the time-line. Fourth, we experiment with an adaptive variant of the semantic class similarity system. The news reflect changes in the real world, and in order to keep up, the system has to change its behavior based on the contents of the news stream. We put forward two strategies for rebuilding the topic representations and report experiment results. We run experiments with three annotated TDT corpora. The use of semantic classes increased the effectiveness of topic tracking by 10-30\% depending on the experimental setup. The gain in spotting new events remained lower, around 3-4\%. The anchoring the text to a time-line based on the temporal expressions gave a further 10\% increase the effectiveness of topic tracking. The gains in detecting new events, again, remained smaller. The adaptive systems did not improve the tracking results.

Measuring and tracking quality factors in Free and Open Source Software projects

Free and Open Source Software (FOSS) has gained increased interest in the computer software industry, but assessing its quality remains a challenge. FOSS development is frequently carried out by globally distributed development teams, and all stages of development are publicly visible. Several product and process-level quality factors can be measured using the public data. This thesis presents a theoretical background for software quality and metrics and their application in a FOSS environment. Information available from FOSS projects in three information spaces are presented, and a quality model suitable for use in a FOSS context is constructed. The model includes both process and product quality metrics, and takes into account the tools and working methods commonly used in FOSS projects. A subset of the constructed quality model is applied to three FOSS projects, highlighting both theoretical and practical concerns in implementing automatic metric collection and analysis. The experiment shows that useful quality information can be extracted from the vast amount of data available. In particular, projects vary in their growth rate, complexity, modularity and team structure.

Unmanned Aerial Vehicles (UAVs) and artificial intelligence revolutionizing wildlife monitoring and conservation

Surveying threatened and invasive species to obtain accurate population estimates is an important but challenging task that requires a considerable investment in time and resources. Estimates using existing ground-based monitoring techniques, such as camera traps and surveys performed on foot, are known to be resource intensive, potentially inaccurate and imprecise, and difficult to validate. Recent developments in unmanned aerial vehicles (UAV), artificial intelligence and miniaturized thermal imaging systems represent a new opportunity for wildlife experts to inexpensively survey relatively large areas. The system presented in this paper includes thermal image acquisition as well as a video processing pipeline to perform object detection, classification and tracking of wildlife in forest or open areas. The system is tested on thermal video data from ground based and test flight footage, and is found to be able to detect all the target wildlife located in the surveyed area. The system is flexible in that the user can readily define the types of objects to classify and the object characteristics that should be considered during classification.

Understanding video based parent training intervention for children with autism

This paper investigates the challenges of delivering parent training intervention for autism over video. We conducted a qualitative field study of an intervention, which is based on a well-established training program for parents of children with autism, called Hanen More Than Words. The study was conducted with a Hanen Certified speech pathologist who delivered video based training to two mothers, each with a son having autism. We conducted observations of 14 sessions of the intervention spanning 3 months along with 3 semi-structured interviews with each participant. We identified different activities that participants performed across different sessions and analysed them based upon their implications on technology. We found that all the participants welcomed video based training but they also faced several difficulties, particularly in establishing rapport with other participants, inviting equal participation, and in observing and providing feedback on parent-child interactions. Finally, we reflect on our findings and motivate further investigations by defining three design sensitivities of Adaptation, Group Participation, and Physical Setup.

In bed with technology: Challenges and opportunities for sleep tracking

In recent years a variety of mobile apps, wearable technologies and embedded systems have emerged that allow individuals to track the amount and the quality of their sleep in their own beds. Despite the widespread adoption of these technologies, little is known about the challenges that current users face in tracking and analysing their sleep. Hence we conducted a qualitative study to examine the practices of current users of sleep tracking technologies and to identify challenges in current practice. Based on data collected from 5 online forums for users of sleep-tracking technologies, we identified 22 different challenges under the following 4 themes: tracking continuity, trust, data manipulation, and data interpretation. Based on these results, we propose 6 design opportunities to assist researchers and practitioners in designing sleep-tracking technologies.

On the integration of self-tracking data amongst quantified self members

Self-tracking, the process of recording one's own behaviours, thoughts and feelings, is a popular approach to enhance one's self-knowledge. While dedicated self-tracking apps and devices support data collection, previous research highlights that the integration of data constitutes a barrier for users. In this study we investigated how members of the Quantified Self movement---early adopters of self-tracking tools---overcome these barriers. We conducted a qualitative analysis of 51 videos of Quantified Self presentations to explore intentions for collecting data, methods for integrating and representing data, and how intentions and methods shaped reflection. The findings highlight two different intentions---striving for self-improvement and curiosity in personal data---which shaped how these users integrated data, i.e. the effort required. Furthermore, we identified three methods for representing data---binary, structured and abstract---which influenced reflection. Binary representations supported reflection-in-action, whereas structured and abstract representations supported iterative processes of data collection, integration and reflection. For people tracking out of curiosity, this iterative engagement with personal data often became an end in itself, rather than a means to achieve a goal. We discuss how these findings contribute to our current understanding of self-tracking amongst Quantified Self members and beyond, and we conclude with directions for future work to support self-trackers with their aspirations.

Implicit and explicit interactions in video mediated collaboration

In this paper we report the results of a study comparing implicit-only and explicit-only interactions in a collaborative, video-mediated task with shared content. Expanding on earlier work which has typically only evaluated how implicit interaction can augment primarily explicit systems, we report issues surrounding control, anxiousness and negotiation in the context of video mediated collaboration. We conclude that implicit interaction has the potential to improve collaborative work, but that there are a multitude of issues that must first be negotiated.

Things you don't want to know about yourself: Ambivalence about tracking and sharing personal information for behaviour change

Technologies that facilitate the collection and sharing of personal information can feed people's desire for enhanced self-knowledge and help them to change their behaviour, yet for various reasons people can also be reluctant to use such technologies. This paper explores this tension through an interview study in the context of smoking cessation. Our findings show that smokers and recent ex-smokers were ambivalent about their behaviour change as well as about collecting personal information through technology and sharing it with other users. We close with a summary of three challenges emerging from such ambivalence and directions to address them.

Human action recognition from video sequences

This PhD research has proposed new machine learning techniques to improve human action recognition based on local features. Several novel video representation and classification techniques have been proposed to increase the performance with lower computational complexity. The major contributions are the construction of new feature representation techniques, based on advanced machine learning techniques such as multiple instance dictionary learning, Latent Dirichlet Allocation (LDA) and Sparse coding. A Binary-tree based classification technique was also proposed to deal with large amounts of action categories. These techniques are not only improving the classification accuracy with constrained computational resources but are also robust to challenging environmental conditions. These developed techniques can be easily extended to a wide range of video applications to provide near real-time performance.

Besides tracking: Simulation of RFID marketing and beyond

This paper asks a new question: how we can use RFID technology in marketing products in supermarkets and how we can measure its performance or ROI (Return-on-Investment). We try to answer the question by proposing a simulation model whereby customers become aware of other customers' real-time shopping behavior and may hence be influenced by their purchases and the levels of purchases. The proposed model is orthogonal to sales model and can have the similar effects: increase in the overall shopping volume. Managers often struggle with the prediction of ROI on purchasing such a technology, this simulation sets to provide them the answers of questions like the percentage of increase in sales given real-time purchase information to other customers. The simulation is also flexible to incorporate any given model of customers' behavior tailored to particular supermarket, settings, events or promotions. The results, although preliminary, are promising to use RFID technology for marketing products in supermarkets and provide several dimensions to look for influencing customers via feedback, real-time marketing, target advertisement and on-demand promotions. Several other parameters have been discussed including the herd behavior, fake customers, privacy, and optimality of sales-price margin and the ROI of investing in RFID technology for marketing purposes. © 2010 Springer Science+Business Media B.V.

Video-stimulated recall in cross-cultural research in education: A case study in Vietnam

This paper examines incorporating video-stimulated recall (VSR) as a data collection technique in cross-cultural research. With VSR, participants are invited to watch video-recordings of particular events that they are involved in; they then recall their thoughts in relation to their observations of their behaviour in relation the event. The research draws on a larger PhD project completed at an Australian university that explored Vietnamese lecturers’ beliefs about learner autonomy. In cross-cultural research using the VSR technique provided significant challenges including time constraints of participants, misunderstandings of the VSR protocol and the possibility of participants’ losing face when reflecting on their teaching episodes. Adaptations to the VSR technique were required to meet the cultural challenges specific to this population, indicating a need for flexibility and awareness of the cultural context for research.

Tracking students through senior schooling: A successful case study

With society now recognizing that senior schooling is about flexibility in credentialing rather than a one-size-fits-all academic education, it has become necessary to track students through numerous pathways. This case study describes how Nambour State High School put into place a senior schooling tracking program which brought about cultural change throughout the school. Using the Tracking and Academic Management Index as a cohort tracking tool, the school has been able to monitor its senior schooling academic and non-academic performance over the past four years. By focusing on the four measures which make up the Index, Nambour State High School was able to demonstrate improved outcomes for all students in their senior school cohort.

Tracking brand positioning for an emerging destination: 2003 to 2015

Few published studies have monitored destination brand image over time. This temporal aspect is an important gap in the literature, given consensus around the role perceptions play in consumers’ decision making, and the ensuing emphasis on imagery in destination branding collateral. Whereas most destination image studies have been a snapshot of perceptions at one point in time, this paper presents findings from a survey implemented four times between 2003 and 2015. Brand image is the core construct in modelling destination branding performance, which has emerged as a relatively new field of research in the past decade. Using the consumer-based brand equity (CBBE) hierarchy, the project has benchmarked and monitored destination brand salience, image and resonance for an emerging regional destination, relative to key competitors, in the domestic Australian market; and the survey instrument has been demonstrated to be reliable in the context of short break holidays by car. What is particularly interesting to date is there has been relatively little change in the market positions of the five destinations, in spite of over a decade of marketing communications by the regional tourism organisations and their stakeholders, and more recently the mass of user-generated travel content on social media. The project didn’t analyse the actual marketing communications for each of the DMOs. Therefore an important implication is that irrespective of the level of marketing undertaken the DMOs seem to have had little control over the perceptions held in their largest market during this time period. Therefore it must be recognised any improvement in perceptions will likely take a long period of time, and so branding needs to be underpinned by a philosophy of a long term financial investment as well as commitment to a consistency of message over time; which given the politics of DMO decision making represents a considerable challenge.

Untitled (after Steven and John)

'Untitled (after Steven and John)' takes inspiration from Spielbergian tracking shots and Baldessarian collages to create ghostly apparitions that explore the affective power of the cinematic close up. By appropriating and obfuscating this common filmic convention, the work investigates the intersubjective potential of the moving image.